Identification and characterization of human DIAPH3 gene in silico

Formin homology proteins with FH1 and FH2 domains are signaling effectors for assembly and polarization of actin filaments. FH1 is the binding domain for Profilin, SRC, EMS1/Cortactin, FNBP1, FNBP2, FNBP3, FNBP4 and WBP4/Fbp21, while FH2 is the actin-filament modification domain. Here, we identified and characterized a novel member of Formin-homology gene family, Diaphanous homology 3 (DIAPH3), by using bioinformatics. DIAPH3 isoform 1, corresponding to 3'-truncated FLJ34705 cDNA and 5'-divergent IMAGE5265490 cDNA, encodes full-length DIAPH3 protein (1112 aa), while DIAPH3 isoform 2, identical to NM_030932.2 cDNA, encodes N-terminally truncated DIAPH3 protein (849 aa). DIAPH3 isoform 1, consisting of exons 1-27, was expressed in lymph node, erythroid progenitor cells as well as in pancreatic cancer. DIAPH3 isoform 2, consisting of exons 1b and 8-27, was expressed in testis. DIAPH3 gene at human chromosome 13q21.2 was found to encode two isoforms due to alternative splicing of the alternative promoter type. Full-length human DIAPH3 protein, consisting of FDD, FH1 and FH2 domains, showed 51.3% total-amino-acid identity with DIAPH1, and 57.3% total-amino-acid identity with DIAPH2. FMNL1/FMNL, FMNL2/FHOD2, FMNL3/WBP3, DAAM1, DAAM2, DIAPH1, DIAPH2 and DIAPH3 were classified as the FDD-type Formin homology proteins, while GRID2IP/Delphilin, FHOD1, Fmn1 and Fmn2 were classified as the non-FDD-type Formin homology proteins. This is the first report on identification and characterization of human DIAPH3 gene.     

Identification and characterization of the human FMN1 gene in silico

Mouse Formin (Fmn1) protein plays a key role in limb morphogenesis. Fmn1 is one of the actin regulators with scaffold function, interacting with Profilin, SRC, EMS1, FNBP1, FNBP2, FNBP3, FNBP4, WBP4 and alpha-catenin. Fmn1, Fmn2, FHOD1, FHOD3, GRID2IP and FHDC1 are non-FDD-type Formin homology proteins, while FMNL1, FMNL2, FMNL3, DIAPH1, DIAPH2, DIAPH3, DAAM1 and DAAM2 are FDD-type Formin homology proteins. Here, we identified the human FMN1 gene by using bioinformatics. The complete coding sequence of human FMN1 cDNA was determined by assembling AC055874.8 genome sequence (nucleotide position 178207-180073), AI040235 EST (complementary sequence for nucleotide position 331-156) and FLJ45135 cDNA (nucleotide position 319-3310). FMN1 isoform 1 (exons 1-18) and FMN isoform 2 (exons 1b and 3-18) were transcribed due to alternative splicing of the alternative promoter type. The FMN1 gene at human chromosome 15q13.3 was located between CKTSF1B1 (Gremlin) and RYR3 genes. The Xenopus fmn1 gene was identified within the Xenopus genome sequence CH216-24N20 (AC147835.1). The FMH1 domain (codon 1-120 of FMN1) and FMH2 domain (codon 683-835 of FMN1) were identified as novel regions conserved among human FMN1, mouse Fmn1, and Xenopus fmn1. The FMH2 domain was almost identical to the alpha-catenin binding domain of mouse Fmn1. Human FMN1 (1419 aa), showing 77.1% total amino-acid identity with mouse Fmn1, was found consisting of FMH1, FMH2, FH1 and FH2 domains. This is the first report on the identification and characterization of the human FMN1 gene as well as the FMH1 and FMH2 domains.     

Identification and characterization of human FHOD3 gene in silico

Formin homology proteins are actin regulators with scaffold function, which are implicated in organogenesis, normal tissue homeostasis, and cancer-cell invasion. FHOD1/FHOS, GRID2IP, Fmn1 and Fmn2 are non-FDD-type Formin homology proteins, while FMNL1, FMNL2/FHOD2, FMNL3, DAAM1, DAAM2, DIAPH1, DIAPH2 and DIAPH3 are FDD-type Formin homology proteins. Here, we identified and characterized FHOD3 (also known as FHOS2), a novel gene homologous to FHOD1, by using bioinformatics. Because FLJ46173, FLJ22297, KIAA1695 and FLJ34580 were partial FHOD3 cDNAs, complete coding sequence of FHOD3 cDNA was determined by assembling nucleotide sequences of FLJ46173 and FLJ22297. FHOD3 gene at human chromosome 18q12.2 was found consisting of at least 25 exons. Exon 11 of FHOD3 gene was spliced out in KIAA1695 cDNA and BF116064 EST, while exon 13 of FHOD3 gene was spliced out in FLJ46173 cDNA. FHOD3 gene encodes at least three isoforms due to alternative splicing of the exon skipping type. FHOD3 and FHOD1 showed 52.1% total-amino-acid identity. Drosophila CG32030 showed 43.9% total-amino-acid identity with human FHOD3, and 39.1% total-amino-acid identity with human FHOD1. FHDHN domain (codon 1-327 of FHOD3) and FHDHC domain (codon 1377-1421 of FHOD3) were identified as the N-terminal conserved region and the juxta C-terminal conserved region, respectively. Human FHOD3, FHOD1 and Drosophila CG32030 were found to share the conserved domain structure consisting of FHDHN, FH1, FH2, and FHDHC domains. This is the first report on the FHOD3 gene as well as on the novel FHDHN and FHDHC domains.     

Identification and characterization of human FHDC1, mouse Fhdc1 and zebrafish fhdc1 genes in silico

Formin homology proteins, implicated in organogenesis and carcinogenesis, are actin regulators with scaffold function. FMNL1, FMNL2, FMNL3, DIAPH1, DIAPH2, DIAPH3, DAAM1 and DAAM2 are FDD-type Formin homology proteins, while FHOD1, FHOD3, GRID2IP, Fmn1 and Fmn2 are non-FDD-type Formin homology proteins. Here, we identified human FHDC1 gene and vertebrate FHDC1 orthologs by using bioinformatics. The complete coding sequence of human FHDC1 cDNA was determined by assembling 3'-recombinated FLJ35083 chimeric cDNA and 5'-truncated KIAA1727 (AB051514.1) partial cDNA. The complete coding sequence of mouse Fhdc1 cDNA was determined by assembling 3'-truncated CD555494 EST and 5'-truncated 6330505N24 (AK031946.1) partial cDNA. The complete coding sequence of zebrafish fhdc1 cDNA was determined by assembling fhdc1 exons within zebrafish genome clone DKEY-4A14 (BX571710.4). FHDC1 gene was located at human chromosome 4q31.3, and Fhdc1 gene at mouse chromosome 3F1. Human FHDC1 (1143 aa) showed 73.3% total amino-acid identity with mouse Fhdc1 (1148 aa), and 43.4% total amino-acid identity with zebrafish Fhdc1 (1165 aa). FDCH1-FDCH5 domains were identified as novel conserved regions among vertebrate FHDC1 orthologs. Human FHDC1, mouse Fhdc1, and zebrafish Fhdc1 were non-FDD-type Formin homology proteins with FH1 and FH2 domains in the N-terminal part as well as with FDCH1, FDCH2, FDCH3, FDCH4, and FDCH5 domains in the C-terminal part. This is the first report on the identification and characterization of the human FHDC1, mouse Fhdc1 and zebrafish fhdc1 genes.     

Identification and characterization of human GRID2IP gene and rat Grid2ip gene in silico

Formin-homology proteins are implicated in the cell polarity control through the assembly of specific actin structures. FMNL1/KW-13/FMNL, FMNL2/KIAA1902/FHOD2, FMNL3/KIAA2014, DAAM1, DAAM2, DIAPH1 and DIAPH2 are Formin-homology proteins with the FDD domain, while Fmn1, Fmn2, FHOD1 and Grid2ip/Delphilin are Formin-homology proteins without the FDD domain. Mouse Grid2ip links glutamate receptor delta2 subunit with actin cytoskeleton and various signaling molecules. Here, we identified and characterized human GRID2IP gene as well as rat Grid2ip gene by using bioinformatics. Human GRID2IP gene was identified within human genome sequence CTD-2195F21 (AC072052.6). Human GRID2IP gene, consisting of 21 exons, was mapped to human chromosome 7p22.1. Rat Grid2ip gene, consisting of 21 exons, was identified within rat genome sequence CH230-82F18 (AC126572.3). Human GRID2IP (1020 aa) showed 91.7% total-amino-acid identity with rat Grid2ip (1024 aa), and 92.7% total-amino-acid identity with mouse Grid2ip. Human GRID2IP protein was found to consist of PDZ domain (codon 94-166), GRCAH domain (codon 204-269), FH1 domain (codon 559-621), and FH2 domain (codon 640-1005). GRCAH domain identified in this study was conserved among mammalian GRID2IP orthologs and mammalian CIP98/KIAA1526 orthologs. This is the first report on comprehensive characterization of human GRID2IP gene as well as on identification of GRCAH domain.     

Identification and characterization of human FNBP1L gene in silico

FNBP1/FBP17/Rapostlin and TRIP10/CIP4 are structurally related microtubule-binding proteins involved in the regulation of cell shape, polarity, motility, and signal transduction. Here, we identified and characterized the FNBP1-like (FNBP1L) gene by using bioinformatics. Human FLJ20275 (NM_017737.1) and mouse 2610318I01Rik (NM_153118.1) were 5'-truncated partial cDNAs derived from human FNBP1L gene and mouse Fnbp1l gene, respectively. Exons 1-7 of FNBP1L gene were located within human genome sequence AL512651.13, and exons 7-15 within AL109613.11. Complete coding sequence of FNBP1L was determined in silico by assembling nucleotide sequences of FNBP1L exons. FNBP1L (547 aa) showed 59.4 and 55.4% total-amino-acid identity with FNBP1 and TRIP10, respectively. FNBP1L, FNBP1 and TRIP10 shared the common domain structure, consisting of FCH, FBH, HR1 and SH3 domains. FCH domain of FNBP1 family proteins is the microtubule-binding domain. HR1 (also known as antiparallel coiled-coil finger) is the binding domain for Rho family proteins, such as ARHN/RhoN and CDC42. SH3 domain of FNBP1 family proteins interacts with proline-rich region of Formin and WASP family proteins. FNBP1L gene was linked to SH2D3B/BCAR3 gene in tail-to-tail manner with an interval less than 8 kb within the human genome. FNBP1L-SH2D3B locus at human chromosome 1p22.1 was paralogous to GPR108-TRIP10-SH2D3A locus at 19p13.3 and GPR107-FNBP1-SH2D3C locus at 9q34.11-q34.13. This is the first report on comprehensive characterization of FNBP1L, which is predicted to function as a scaffold protein for microtubule, Rho family proteins, Formin-homology proteins and WAPS family proteins.     

Identification and characterization of human FCHSD1 and FCHSD2 genes in silico

FNBP1 and FNBP2 are SH3-type Formin-binding proteins. FNBP1 consists of FCH, FBH, HR1 and SH3 domains, while FNBP2 consists of FCH, FBH, RhoGAP and SH3 domains. Here, we identified novel genes FCHSD1 and FCHSD2, which were distantly related to FNBP1 and FNBP2. FCHSD1 and FCHSD2 genes with conserved exon-intron structure were located at human chromosome 5q31.3 and 11q13.4, respectively. Complete coding sequence of human FCHSD1 was derived from FLJ00007 (NM_033449.1) cDNA. KIAA0769 (NM_014824.1), encoding N-terminally truncated 684-aa protein, was an aberrant human FCHSD2 cDNA with a frame shift due to skipping of 98-bp exon 2. Complete coding sequence of human FCHSD2 cDNA was determined by assembling CF995054 EST and KIAA0769 cDNA. A030002D08Rik (NM_175684.3) was the representative mouse Fchsd1 cDNA, while BC034086 (NM_199012.1) was a variant mouse Fchsd2 cDNA with an insertion of 72-bp additional exon. CG4684 was the Drosophila homolog of mammalian FCHSD family genes. Human FCHSD1 (690 aa) showed 41.7% total-amino-acid identity with human FCHSD2 (740 aa), and 91.0% total-amino-acid identity with mouse Fchsd1. Human FCHSD2 showed 96.5% total-amino-acid identity with mouse Fchsd2. Mammalian FCHSD family proteins shared the common domain structure consisting of FCH, FBH, two SH3 and C-terminal Proline-rich domains. FCHSD family proteins (FCHSD1 and FCHSD2), FNBP1 family proteins (FNBP1, FNBP1L and TRIP10/CIP4) and FNBP2 family proteins (FNBP2, ARHGAP13/SRGAP1, ARHGAP14/SRGAP2 and ARHGAP4) were found constituting the FCFBS superfamily characterized by FCH, FBH and SH3 domains. This is the first report on identification and characterization of the FCHSD family genes.     

Crystal structure of human DAAM1 formin homology 2 domain

Reorganization of the actin filament is an essential process for cell motility, cell-cell attachment and intracellular transport. Formin proteins promote nucleation and elongation of the actin filament, and thus are key regulators for this process. The formin homology 2 (FH2) domain forms a head-to-tail ring-shaped dimer, and processively moves towards the barbed end. Dishevelled-associated activator of morphogenesis (DAAM) is a Rho-regulated formin implicated in neuronal development. Here, we present the crystal structure of human DAAM1 FH2 dimer at 2.8 A resolution. This is the first dimeric structure of the mammalian formin. The core structure of human DAAM1 is similar to those of mouse mDia1 and yeast Bni1p, whereas the orientations of the FH2 dimeric rings are different between human DAAM1 and yeast Bni1p, despite their similar dimer interactions. This difference supports the previous prediction that the dimer architecture of the formin is highly flexible in the actin-free state. The results of the actin assembly assays using the DAAM1 mutants demonstrated that the length of the linker connecting the N-terminal domain and the core region is crucial for the activity.     

Identification and characterization of rat Dact1 and Dact2 genes in silico

DACT1 (DAPPER1), Frizzled receptors, MUSK receptor, VANGL1, VANGL2, PRICKLE1, PRICKLE2, DAAM1, Casein kinases, MARK3 (PAR1), PP2C, AXIN1, AXIN2, NKD1, NKD2, FRAT1, FRAT2 and CXXC4 are WNT signaling molecules associating with Dishevelled family proteins. Human DACT1 is the ortholog of Xenopus Dapper and Frodo, and human DACT2 (DAPPER2) is the paralog of human DACT1. Here, we identified and characterized rat Dact1 (Dapper1) and Dact2 (Dapper2) genes by using bioinformatics. Rat Dact1 gene, consisting of four exons, was located within AC136677.3 genome sequence. Rat Dact2 gene, consisting of four exons, was located within AC139434.3 genome sequence. Dact1 was mapped to rat chromosome 6q24, and Dact2 gene to rat chromosome 1q12. Rat Dact1 (778 aa) showed 93.7, 82.9, 60.3, 58.7 and 48.6% total-amino-acid identity with mouse Dact1, human DACT1, Xenopus Dapper, Xenopus Frodo and zebrafish dact1, respectively. Rat Dact2 (768 aa) showed 86.6, 59.6 and 38.3% total-amino-acid identity with mouse Dact2, human DACT2 and zebrafish dact2, respectively. Dact1 orthologs were more evolutionarily conserved than Dact2 orthologs. Seven DAPH domains (DAPH1-DAPH7), originally identified as the regions conserved between human DACT1 and DACT2, were conserved among mammalian Dact1 orthologs and Dact2 orthologs. DAPH2 domain, corresponding to the Leucine zipper motif, was located within the coiled-coil region. DAPH3 domain was the Serine rich region. DAPH7 domain was the C-terminal PDZ binding region. This is the first report on the rat Dact1 and Dact2 genes.     

FNBP2 gene on human chromosome 1q32.1 encodes ARHGAP family protein with FCH,FBH, RhoGAP and SH3 domains

Parallel analyses of DNA copy number and mRNA expression level by using microarray measurements have been successfully applied for genome-wide screening of proto-oncogenes and tumor suppressor genes. The uncharacterized KIAA0456 cDNA was reported amplified and overexpressed in human breast cancer cell lines UACC812 and ZR-75-1. Here, we characterized the gene corresponding to KIAA0456 cDNA by using bioinformatics. KIAA0456 cDNA was found derived from the FNBP2 gene, consisting of 22 exons. FNBP2 gene was linked to IKBKE and NORE1 genes on human chromosome 1q32.1. FNBP2 mRNA was expressed in melanoma, germ cell tumors, chondrosarcoma and retinoblastoma. ARHGAP13/SRGAP1, ARHGAP14/SRGAP2 and ARHGAP4 cDNAs were homologous to FNBP2/KIAA0456 cDNA. KIAA1304 cDNA was a splicing variant derived from the ARHGAP13 gene, and the nucleotide sequence of representative ARHGAP13 cDNA was determined by assembling EST BU520980 and KIAA1304 cDNA. FNBP2 (1071 aa), ARHGAP13 (1062 aa), ARHGAP14 (1099 aa) and ARHGAP4 (946 aa) constitute the FNBP2 family characterized by FCH, RhoGAP and SH3 domains. The region corresponding to codon 227-345 of FNBP2 was conserved among FNBP2 family proteins as well as FNBP1 and TRIP10 proteins. Because FNBP2 and FNBP1 are formin-binding proteins, the region corresponding to codon 227-345 of FNBP2 was designated FNBP2-FNBP1 homologous (FBH) domain. FNBP2 family proteins consist of FCH, FBH, RhoGAP and SH3 domains, while FNBP1 family proteins (FNBP1 and TRIP10) consist of FCH, FBH and SH3 domains. This is the first report on comprehensive characterization of FNBP2 gene as well as on identification of the FBH domain.     

Identification and characterization of ARHGAP27 gene in silico

ARHGAP1, ARHGAP2, ARHGAP3, ARHGAP4, ARHGAP5, ARHGAP6, ARHGAP7 (DLC1), ARHGAP8, ARHGAP9, ARHGAP10, ARHGAP12, ARHGAP13 (SRGAP1), ARHGAP14 (SRGAP2), ARHGAP15, ARHGAP17 (RICH1), ARHGAP18, ARHGAP19, ARHGAP20, ARHGAP21, ARHGAP22, ARHGAP23, ARHGAP24, ARHGAP25, ARHGAP26, STARD13 (DLC2), HA-1, GMIP, PARG1, RACGAP1, PIK3R1, PIK3R2, and FNBP2 genes encode Rho/Rac/Cdc42-like GTPase activating (RhoGAP) proteins. Here, we characterized human ARHGAP27 gene by using bioinformatics. Complete coding sequence of ARHGAP27 isoform 1, encoding a full-length 889-aa protein, was determined by assembling exon 1 (nucleotide position 143440-144096 of AC091132.16) and most part of FLJ43547 cDNA (nucleotide position 69-3628 of AK125535.1). Complete coding sequence of ARHGAP27 isoform 2, encoding an N-terminally truncated 548-aa protein, was derived from FLJ43547 cDNA. ARHGAP27 isoform 1 consists of exons 1-17, while ARHGAP27 isoform 2 consists of exons 1B, and 2-17. ARHGAP27 gene encoded two isoforms due to alternative splicing of alternative promoter type. ARHGAP27 mRNA was expressed in germinal center B cell, spleen, chronic lymphocytic leukemia, pancreatic cancer, and lung cancer. LOC303583 (NM_ 198759.1) was the representative rat Arhgap27 cDNA. Human ARHGAP27 showed 84.3% total-amino-acid identity with rat Arhgap27, and 39.0% total-amino-acid identity with human ARHGAP12. ARHGAP27 and ARHGAP12 shared the common-domain structure, consisting of SH3, WW, PH, and RhoGAP domains. ARHGAP27 gene was located at human chromosome 17q21, while ARHGAP12 gene was located at human chromosome 10p11. ARHGAP family genes are cancer-associated genes, because their genetic alterations lead to carcinogenesis through the dysregulation of Rho/Rac/ Cdc42-like GTPases. This is the first report on identification and characterization of the ARHGAP27 gene.     

Identification and characterization of human MPP7 gene and mouse Mpp7 gene in silico

Drosophila Crumbs (Crb), Stardust (Sdt), Discs large (Dlg), Scribble (Scrib) and Lethal giant larvae (Lgl) are involved in the establishment and the maintenance of apicobasal polarity in epithelial tissues. Because epithelial polarity is disrupted in tumors, human homologs of Drosophila crb, sdt, dlg, scrib, and lgl are potential cancer-associated genes. MPP1/EMP55, MPP2, MPP3, MPP4, MPP5/PALS1 and MPP6/PALS2 genes are human homologs of Drosoplila sdt. Here, we identified and characterized a novel member of MPP gene family, MPP7, by using bioinformatics. Uncharacterized FLJ32798 cDNAs (BC038105 and AK057360) were derived from human MPP7 gene. BC038105 was a representative MPP7 cDNA, while AK057360 was an aberrant MPP7 cDNA with a frame shift. Human MPP7 mRNA was expressed in placenta, brain, testis as well as in uterus tumor, bladder tumor, and lymphoma. Microsatellite marker D10S588, linked to IDDM and hereditary thrombocytopenia, was located within the MPP7 gene at human chromosome 10p12.1. Nucleotide sequence of mouse Mpp7 cDNA was determined in silico by assembling 3'-truncated cDNA AK078849, genome clone RP24-255J24, and EST AV260217. Human MPP7 showed 92.9% total-amino-acid identity with mouse Mpp7, and 75.7% total-amino-acid identity with zebrafish humpback. MPP7 orthologs were MAGUK proteins with two L27 domains, PDZ domain, SH3 domain, and GuKc domain. MPP7 was most related to MPP3 among MPP family members, functioning as adopter molecules assembling Crb homologs (CRB1, CRB3), Dlt homologs (INADL/PATJ, MPDZ/MUPP1), and Lin-7 homologs (LIN7A, LIN7B, LIN7C). This is the first report on identification and characterization of human MPP7 and mouse Mpp7 genes.     

Comparative genomics on Norrie disease gene

DAND1 (NBL1), DAND2 (CKTSF1B1 or GREM1 or GREMLIN), DAND3 (CKTSF1B2 or GREM2 or PRDC), DAND4 (CER1), DAND5 (CKTSF1B3 or GREM3 or DANTE), MUC2, MUC5AC, MUC5B, MUC6, MUC19, WISP1, WISP2, WISP3, VWF, NOV and Norrie disease (NDP or NORRIN) genes encode proteins with cysteine knot domain. Cysteine-knot superfamily proteins regulate ligand-receptor interactions for a variety of signaling pathways implicated in embryogenesis, homeostasis, and carcinogenesis. Although Ndp is unrelated to Wnt family members, Ndp is claimed to function as a ligand for Fzd4. Here, we identified and characterized rat Ndp, cow Ndp, chicken ndp and zebrafish ndp genes by using bioinformatics. Rat Ndp gene, consisting of three exons, was located within AC105563.4 genome sequence. Cow Ndp and chicken ndp complete CDS were derived from CB467544.1 EST and BX932859.2 cDNA, respectively. Zebrafish ndp gene was located within BX572627.5 genome sequence. Rat Ndp (131 aa) was a secreted protein with C-terminal cysteine knot-like (CTCK) domain. Rat Ndp showed 100, 96.9, 95.4, 87.8 and 66.4 total-amino-acid identity with mouse Ndp, cow Ndp, human NDP, chicken ndp and zebrafish ndp, respectively. Exon-intron structure of mammalian Ndp orthologs was well conserved. FOXA2, CUTL1 (CCAAT displacement protein), LMO2, CEBPA (C/EBPalpha)-binding sites and triple POU2F1 (OCT1)-binding sites were conserved among promoters of mammalian Ndp orthologs.     

Comparative genomics on FGF8, FGF17, and FGF18 orthologs

FGF and WNT signaling pathways network together during embryogenesis and carcinogenesis. Among 22 FGF family members within human and rodents genomes, FGF20 orthologs are evolutionarily conserved targets of the WNT/beta-catenin signaling pathway. FGF8, FGF17, and FGF18 constitute one of FGF subfamilies. Here, comparative proteomics and comparative genomics analyses on FGF8, FGF17, and FGF18 orthologs were performed. Rat Fgf8 and Fgf17 genes, consisting of five exons, were located within AC096326.7 and AC097410.12 genome sequences, respectively. FGF8, FGF17, and FGF18 orthologs were FGF family members with the N-terminal signal peptide. Human FGF8 isoform F showed 90.6% total-amino-acid identity with rat Fgf8 (268 aa). Human FGF17 showed 98.6% total-amino-acid identity with rat Fgf17 (216 aa). Human FGF18 also showed 98.6 total-amino-acid identity with rat Fgf18. FBXW1 (betaTRCP1 or BTRC1)-FGF8-NPM3 locus at human chromosome 10q24.32, FBXW11 (betaTRCP2 or BTRC2)-FGF18-NPM1 locus at human chromosome 5q35.1, and FGF17-NPM2 locus at human chromosome 8p21.3 were paralogous regions within the human genome. FGF8 mRNA was expressed in DMSO-treated embryonic stem (ES) cells. FGF17 mRNA was expressed in ES cells differentiated to an early endodermal phenotype. FGF18 mRNA was expressed in fetal lung, fetal heart, lung carcinoid, colorectal cancer, and ovarian cancer. FGF18 promoter with double TCF/LEF binding sites rather than FGF8 promoter and FGF17 promoter was more conserved between human and rodents. These facts indicate that FGF18 orthologs were evolutionarily conserved targets of the WNT/beta-catenin signaling pathway.     

Identification and characterization of human PDZRN4L gene and mouse Pdzrn4l gene in silico

LNX, functioning as E3 ubiquitin ligase for NUMB, is implicated in the cell fate determination through the inhibition of Notch signaling. LNX, PDZRN1 (LNX2), PDZRN3 (LNX3 or SEMCAP3) and PDZRN4 (LNX4 or SEMCAP3L) constitute the LNX (PDZRN) family. PDZRN4 gene encodes 2 isoforms due to alternative splicing. PDZRN4 consists of RING, 2 PDZ, PR34H1 and PR34H2 domains, and PDZRN4S consists of PDZ, PR34H1 and PR34H2 domains. Here, we identified novel PDZRN4-related genes by using bioinformatics. FLJ45072 (AK127016.1) and KIAA1444 (NM_032512.1) cDNAs were derived from human PDZRN4L (also known as PDZRN5, LNX4L, or LNX5) gene. FLJ45072 was the representative PDZRN4L cDNA, while KIAA1444 was a 5'-truncated partial PDZRN4L cDNA. MGC67228 (BC056462.1) and B230341P03 (AK046101.1) cDNAs were derived from mouse Pdzrn4l gene. MGC67228 was a 5'-truncated partial Pdzrn4l cDNA, while B230341P03 was an aberrant Pdzrn4l cDNA with exon skipping and insertions within the coding region. PDZRN4L gene, consisting of at least 8 exons, was located at human chromosome Xq28. Exons 1-8 of PDZRN4L gene corresponded to exons 1b, 4-10 of PDZRN4 gene. Because the regions corresponding to exons 1-3 of PDZRN4 gene were not identified within human genome sequences around the PDZRN4L gene, PDZRN4L isoform with RING finger domain was not identified. Human PDZRN4L (769 aa) showed 93.0% total-amino-acid identity with mouse Pdzrn4l (772 aa), and 49.9% total-amino-acid identity with human PDZRN4S. PDZ, PR34H1 and PR34H2 domains were conserved between PDZRN4L and PDZRN4S. This is the first report on human PDZRN4L and mouse Pdzrn4l genes.     

Molecular cloning and characterization of mouse Wnt14b, clustered with mouse Wnt3 in mouse chromosome 11.

WNTs are a family of secreted-type glycoproteins implicated in embryogenesis and carcinogenesis. We have previously cloned and characterized human WNT2B/WNT13, WNT3, WNT3A, WNT5B, WNT6, WNT7B, WNT8A, WNT8B, WNT10A, WNT10B, WNT11, WNT14, and WNT14B/WNT15. WNT14B gene is clustered with WNT3 gene in human chromosome 17q21, and mRNA expression of WNT14B is significantly up-regulated by retinoic acid during the early phase of neuronal differentiation in human NT2 cells. Here, we identified mouse Wnt14b gene fragments in mouse genome draft sequence AL596108.5 by using bioinformatics, and isolated mouse Wnt14b cDNAs by using cDNA-PCR. Mouse Wnt14b was found to encode a 359-amino-acid WNT family protein with the N-terminal signal peptide, an N-linked glycosylation site, and 24 conserved cysteine residues. Mouse Wnt14b showed 92.5% total-amino-acid identity with human WNT14B, and 64.2% total-amino-acid identity with human WNT14. Mouse Wnt14b gene, consisting of 4 exons, was clustered with mouse Wnt3 gene in mouse chromosome 11. Mouse Wnt14b mRNA was relatively highly expressed in 17-day embryo, and also expressed in adult brain, kidney, liver, 7-day embryo, and 11-day embryo. This is the first report on molecular cloning and characterization of mouse Wnt14b.     

Comparative genomics on Fzd7 orthologs

WNT signaling pathway networks with Hedgehog and Notch signaling pathways during carcinogenesis and embryogenesis. FZD7 is up-regulated in gastric cancer, esophageal cancer, and hepatocellular carcinoma (HCC). Here we identified and characterized rat Fzd7 gene by using bioinformatics. Rat Fzd7 gene was identified within AC136379.2 genome sequence. The 5'-flanking region and exonic region were well conserved among mammalian Fzd7 orthologs. Nucleotide position 153000-152216 of AC136379.2 genome sequence was identified as the evolutionarily conserved promoter region of rat Fzd7 gene, and nucleotide position 2273-3046 of AC069148.6 genome sequence as the evolutionarily conserved promoter region of human FZD7 gene. Match program revealed that PAX4-binding site was conserved among rat Fzd7, mouse Fzd7 and human FZD7 promoters. Rat Fzd7 (572 aa) was a seven-transmembrane-type Wnt receptor, which showed 99.3, 96.9, 87.4, 85.5, 79.5 and 79.0% total-amino-acid identity with mouse Fzd7, human FZD7, chicken fzd7, Xenopus fzd7, zebrafish fzd7a and fzd7b, respectively. Frizzled (Fz) domain within the N-terminal extracellular region, Leucine zipper motif around the fifth transmembrane (TM5) region, Dishevelled (Dvl)- and Magi3-binding motifs within the C-terminal cytoplasmic region were conserved among vertebrate Fzd7 orthologs. Leucine zipper motif around the TM5 region of Fzd7 orthologs was disrupted in FzE3 aberrant cDNA due to multiple cloning artifacts or sequencing errors. These facts indicate that experimental data obtained by using FzE3 cDNA do not always reflect the functions of Fzd7 orthologs.